Emotions and cost/benefit assessment The role of shame and self-esteem in risk taking

Ethnographic and experimental findings indicate that emotions influence decision making in a number of ways, including the weighting of cost/benefit assessments. One pair of emotions, Shame and Pride, bias such assessments in the service of both rank-striving and approvalseeking behaviors. The frequency with which these emotions influence decision making is determined by self-esteem, a mechanism which sums events to date. This allows for the adjustment of risk taking choices in light of future prospects. Studies of humans and nonhuman primates suggest that serotonin, a neurotransmitter, plays a key role in such a system. In the environment in which humans evolved, this patterned attunement of cost/benefit assessments would have optimized fitness. INTRODUCTION The world is not intrinsically divided into options. Instead, an organism must parse stimuli in such a way as to define events, and thus to define choices. Because decision making is logically prior to action, we can expect that natural and sexual selection will have acted to create in organisms the capacity to divide up the world in this manner in order to facilitate adaptive behavior (Tooby and Cosmides 1990). Finer parsing of the environment produces greater flexibility and complexity of behavior: Whereas a single-celled organism may chunk the world along only a few axes (warm versus cold, edible versus toxic, etc.), and may possess only a few responses (approach versus avoid, etc.), creatures as complex as mammals divide the world into many more categories, and possess a much wider range of responses. Such complexity introduces still more complexity, as meta-decisions must then be addressed, i.e., it becomes necessary to prioritize decisions to be made. Humans, and probably other mammals, (and perhaps other vertebrates), seem to manage these various demands not through brute information-processing power, but rather through discrete decision making apparatai which shape and constrain the tasks at hand. Specifically, it appears that emotions are integral parts of decision making mechanisms which parse the world into decisions, prioritize some decisions over others, and weight particular options within decisions (ibid.). Any decision embodies an assessment of the future, for the value of a given option is contingent on what the organism’s circumstances are likely to be the next moment, day, year, or decade. Organisms thus require a means of prognostication. The most reliable way of predicting the future is to extrapolate from the past. Organisms therefore probably possess mechanisms which aggregate experience and, in so doing, influence the proximate mechanisms of decision making. In this paper I will argue that a pair of emotions, shame and pride, importantly influence decision making. These emotions, in turn, are linked to self-esteem, a mechanism which sums experience to date (where the nature of that experience is itself defined in emotional terms). The result is a decision making system which adjusts risk taking behavior in accord with probable future opportunities, where such opportunities are predicted on the basis of past events. DRIVES AND EMOTIONS Observing patterned, species-typical behavior which appears strongly goal-oriented, investigators have often posited the existence of discrete drives. However, “drive” is simply a label for a black box -the mechanism is defined wholly in terms of its output. Moreover, when we examine actual human behavior, we find that informants often explain actions largely in terms of emotions. Individual emotion events, while importantly shaped by both idiosyncratic past experience and the cultural context, are nonetheless constructed upon panhuman capacities and proclivities to experience particular reactions to particular classes of events. Presumably, such marked and pervasive features evolved because they provided adaptive advantages to those who possessed them (Tooby and Cosmides 1990). How might emotions be advantageous? First, emotions inform the individual about how she currently stands in relation to the world. Each emotion is an integral part of a system which parses the world into decision categories by ‘telling’ the individual about particular types of relationships with the world (ibid.). Fear, for example, is a way of ‘knowing’ that the environment is threatening to the individual -the emotion, elicited by particular stimuli in the environment, tags a particular type of event in the world, causing it to be salient. Second, emotions heighten and prolong monitoring of stimuli relevant to the given type of event, probably at the expense of other stimuli -fear maintains the attention directed at potentially threatening stimuli (reviewed in Mineka and Sutton 1992). Third, emotions influence the recall of stored information, as memories relevant to the given type of event become more accessible, while other information is more difficult to recall – during an exam, a fearful student may be able to remember how one can escape from frightening situations, but he may be unable to recall the course material. Fourth, emotions proportionately influence meta-decision making (Tooby and Cosmides 1990) -the experience of intense fear prioritizes decisions concerning the eliciting threat ahead of any other decisions, such as what to have for dinner. Fifth, emotions help to constrain the available choices to those which will address the eliciting event (ibid.) -fear may or may not assist an individual in deciding which way to flee (see endnote), but it clearly inclines the individual to flee, rather than to engage in any of an infinite variety of other possible behaviors. Sixth, because they are distinctly rewarding or aversive, emotions direct decision making in a number of ways. In the simplest case, individuals experience an emotion and then act to either terminate or extend the circumstances which elicited it, depending upon the emotion’s hedonic value. More complexly, as a consequence of past experience, individuals actively seek to either create or avoid the circumstances which elicit a given emotion. Furthermore, emotions are sometimes paired such that situation X elicits a rewarding emotion, while situation NOT-X elicits an aversive emotion. This redundancy increases the likelihood that an actor will seek out or avoid particular types of situations, that is, it lends robustness to the ordered nature of decisions. Seventh, as will be discussed at length below, emotions may influence decision making by affecting the relative salience or weight of costs versus benefits. Finally, because multiple emotions may be elicited simultaneously, emotions’ influence on decision making is sometimes the product of the interaction among emotions. When two emotions having antithetical features, such as fear and love, are elicited simultaneously, indecision may occur (until such time as one emotion is more strongly elicited). Conversely, when two emotions having complementary features, such as fear and disgust, are elicited simultaneously, the result may be an additive influence on decision making. In sum, evolution appears to have patterned decision making, and hence behavior, by shaping the eliciting conditions, cognitive consequences, action tendencies, hedonic values, and biasing effects of specific emotions. We can therefore think of “drives” as consisting of evolved sets of emotions which work together to produce particular outcomes in particular circumstances. SHAME AND RISK All decisions involve a weighing up of the relative costs and benefits of different courses of action. As noted above, emotions influence decision making in part by biasing some outcomes over others. Sometimes, such biasing is in keeping with a common-sensical understanding of the ‘sensible’ choice in the given situation -the benefits of running away from a large predator obviously far outweigh the costs, and hence we are content to describe the decision to flee as ‘sensible’ even if, in fact, it was the product of an emotional reaction rather than a conscious calculation. Often, however, common-sensical assessments view emotional decisions as ‘senseless,’ particularly when the potential costs, or risks (the costs considered in light of the likelihood that they will be incurred), seem to outweigh the potential benefits. Nowhere is this more true than in instances of spontaneous violence. Below is a case drawn from my anthropological fieldwork in Bengkulu, southwestern Sumatra: Rustam and his girlfriend left the party around 11:30 p.m. They found a minibus and, after bargaining with the driver and his assistant, they set off. As they neared Rustam’s village, a disagreement arose over the fare. The discrepancy was 50 rupiah, the price of a single cracker at a village shop. Words were passed. A fight began. The two young men stabbed Rustam 14 times and dumped him by the side of the highway. He died as his girlfriend knelt wailing at his side. The next day, after the killers had been arrested, the body had been buried, and the stain by the highway had been washed away, people gathered in the shops. “They must’ve been possessed by the Devil!” several old men said, “Imagine, killing somebody over 50 rupiah. Only a madman would do such a thing!” But the young men knew that possession and insanity were only metaphors. “It was all because of malu,” one said, and others nodded. “No one wants to be malu in front of a girl.” Viewed cross-culturally, Rustam’s death was a prototypical murder. Violent interactions frequently involve young men, and frequently begin as altercations over trivial points, often with an audience looking on (Daly and Wilson 1988). From a common-sensical position, such events seem notably senseless, as the contested point appears insignificant relative to the potential costs. However, as Daly and Wilson have compellingly argued, it is likely that young men possess the propensity to make such choices because, in our evolutionary past, such reactions would have allowed a maturing male with little status or influence to rapidly establish a reputation as someone to be taken seriously in the highly competitive male social arena. Hence, if utility is defined in terms of inclusive fitness, while such a choice may or may not be rational now, it was rational in the evolutionary context in which it arose (cf. Hammerstein, this volume) -given high variance in male reproductive success, risk taking made sense for young males with uncertain futures. Note that the above explanation operates at the ultimate level -we have an account of why a predisposition exists, but not of how it operates. To gain insight into the proximate workings of the system, we return to informants’ analyses of events. In explaining the combatants’ behavior, villagers made frequent reference to malu, an aversive, shame-like emotion. Apparently, the experience of this emotion (combined with the prospect of its intensification) colored the participants’ assessments of the costs and benefits of escalation. In order to understand why malu had this effect, we must explicate this emotion more fully. It is possible to describe an emotion in terms of what I call its ‘logic,’ that is, the abstracted set of conditions wherein it is experienced. The Bengkulu emotion malu, like the English emotion shame, is characterized by the following 6-point logic: 1) Ego violates a norm 2) Ego is aware of his failure 3) an Other is also aware of Ego’s failure 4) Ego is aware of the Other’s knowledge 5) the Other displays hostility and revulsion towards Ego OR Ego assumes that Other experiences hostility and revulsion towards Ego 6) as a consequence of the above, Ego experiences an aversive emotion Conversely, bangga, the opposite of malu, is characterized by the same logic as the English emotion pride, the opposite of shame: 1) Ego successfully fulfills a norm 2) Ego is aware of her success 3) an Other is also aware of Ego’s success 4) Ego is aware of the Other’s knowledge 5) Other displays towards Ego either i) a positive appraisal and affection, or ii) a positive appraisal and hostility OR Ego assumes that Other experiences (i) or (ii) towards Ego 6) as a consequence of the above, Ego experiences a pleasurable emotion At first glance it thus appears that malu and shame are isomorphic, as are bangga and pride. Moreover, cross-cultural comparison (see Fessler 1999) reveals that disparate cultures possess the same pair of opposing 6-point logic emotions, suggesting that a shame-like emotion and a pride-like emotion are probable human universals. However, this observation does not in itself shed light on how a shame-like emotion could have led to Rustam’s murder, since the fight did not revolve around any clear violation of a norm. To understand the connection between malu and violence, we must explore this emotion still further. In addition to being characterized by the 6-point logics, malu and its opposite, bangga, are also characterized by simpler 3-point logics, as follows: 1) Ego assesses an Other as significantly more important than Ego 2) Ego must interact with the Other in a situation in which the discrepancy between Ego and the Other is salient for Ego 3) as a consequence of the above, Ego experiences an aversive emotion and 1) Ego assesses an Other as significantly less important than Ego 2) Ego must interact with the Other in a situation in which the discrepancy between Ego and the Other is salient for Ego 3) as a consequence of the above, Ego experiences a pleasurable emotion Moreover, it is clear that these are not cases of simple polysemy, as the same display and outcome behaviors, familiar to English speakers, are associated with each term whether it is in regard to the 6-point or the 3-point logic: